ALCF Announces Online AI Testbed Training Workshops

The Argonne Leadership Computing Facility has announced a series of training workshops to introduce users to the novel AI accelerators deployed at the ALCF AI Testbed. The four individual workshops will introduce participants to the architecture and software of the SambaNova DataScale SN30, the Cerebras CS-2 system, the Graphcore Bow Pod system, and the GroqRack […]

Groq Acquires Definitive Intelligence to Launch GroqCloud

MOUNTAIN VIEW, Calif., March 1, 2024 — AI technology company Groq has acquired Definitive Intelligence to launch GroqCloud, “a new developer playground with fully integrated documentation, code samples, and self-serve access.” Definitive Intelligence Co-founder and CEO Sunny Madra will lead the new GroqCloud business unit with the goal of expanding access to the Groq LPU Inference […]

ALCF: Groq AI Online Workshop, Dec. 6-7

The Argonne Leadership Computing Facility will host an online workshop to learn about the ALCF’s new GroqRack system on Monday and Tuesday, Dec. 6 and 7 from 1:30 to 5 pm Central Time. To register go here. The ALCF said the Groq machine’s AI inference capabilities can assist users in making predictions, discovering patterns in […]

@HPCpodcast: Industry Analyst Karl Freund Compares and Contrasts the Chips Driving the AI Revolution

Few people can talk in such an engaging way about the chips driving the AI revolution as Karl Freund, founder and principal analyst at Cambrian-AI Research. He joins us to discuss the, well, “Cambrian explosion” we’re witnessing in AI chips, the general state of the AI semiconductor….

ALCF Deploys Testbed to Advance AI for Science

The Argonne Leadership Computing Facility (ALCF) is building a testbed comprised of advanced artificial intelligence (AI) platforms. Designed to explore the possibilities of high-performance computing (HPC) architectures, the ALCF AI Testbed will enable the facility and its user community to help define the role of AI accelerators in next-generation scientific machine learning. “The testbed combines a […]

Groq AI Chip Benchmarks Leading Performance on ResNet-50 Inference

Today AI chip startup Groq announced that their new Tensor processor has achieved 21,700 inferences per second (IPS) for ResNet-50 v2 inference. Groq’s level of inference performance exceeds that of other commercially available neural network architectures, with throughput that more than doubles the ResNet-50 score of the incumbent GPU-based architecture. ResNet-50 is an inference benchmark for image classification and is often used as a standard for measuring performance of machine learning accelerators.